Model comparison
Gemma 3 27B vs Mixtral 8x22B
Gemma 3 27B is the stronger model overall, scoring 30.8 to 27.1 on the Noometry Index.
Last verified . 21 shared benchmarks.
Google
Mistral AISummary
- They share 21 benchmarks with published results for both. Gemma 3 27B scores higher in 6 categories and Mixtral 8x22B in 3 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemma 3 27B leads 52.5 to 36.9.
- The biggest single-benchmark swing is MATH Level 5: 74% for Gemma 3 27B and 24.2% for Mixtral 8x22B.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $2 / $6 for Mixtral 8x22B.
- Gemma 3 27B accepts more context: 131K tokens versus 64K.
Side by side
| Gemma 3 27B | Mixtral 8x22B | |
|---|---|---|
| Provider | Mistral AI | |
| Noometry Index | 30.8 | 27.1 |
| Released | 2025-03-11 | 2024-04-17 |
| Weights | Open | Open |
| Context window | 131K | 64K |
| Max output | 8K | 64K |
| Input $ / M tokens | $0.08 | $2 |
| Output $ / M tokens | $0.16 | $6 |
| Results tracked | 43 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mixtral 8x22B leads
Gemma 3 27B: 22.5 (#334), Mixtral 8x22B: 24.2 (#329)
| Benchmark | Gemma 3 27B | Mixtral 8x22B |
|---|---|---|
| LMArena Coding | 1322 | 1166 |
| Aider Polyglot | 4.9% | — |
| SciCode | 21.2% | — |
| WeirdML | — | 3.2% |
| BigCodeBench Instruct | — | 40.6% |
| LiveBench Coding | 39.9% | — |
| BigCodeBench Complete | — | 50.2% |
| HumanEval+ | — | 72% |
| MBPP+ | — | 64.3% |
Agentic & Tool Use Gemma 3 27B leads
Gemma 3 27B: 25.1 (#110), Mixtral 8x22B: 23.1 (#127)
| Benchmark | Gemma 3 27B | Mixtral 8x22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 29.5% | — |
| Cybench | — | 7.5% |
Reasoning Mixtral 8x22B leads
Gemma 3 27B: 16.7 (#301), Mixtral 8x22B: 19.9 (#248)
| Benchmark | Gemma 3 27B | Mixtral 8x22B |
|---|---|---|
| LMArena Hard Prompts | 1340 | 1150 |
| DTBench | 52.5% | 55.1% |
| Epoch Capabilities Index | 130.04 | 122.03 |
| Kagi LLM Benchmark | 40.4% | — |
| CritPt | 0% | — |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | 43.8% | — |
| LiveBench Data Analysis | 51.5% | — |
| LMCA | 12.3% | — |
| ForecastBench | — | 56.3 |
| LiveBench | 50% | — |
Math Gemma 3 27B leads
Gemma 3 27B: 25.9 (#265), Mixtral 8x22B: 22.9 (#275)
| Benchmark | Gemma 3 27B | Mixtral 8x22B |
|---|---|---|
| LMArena Math | 1312 | 1184 |
| MATH Level 5 | 74% | 24.2% |
| OTIS Mock AIME 2024-2025 | 22.5% | — |
| Omni-MATH | — | 16.3% |
| LiveBench Math | 55.4% | — |
Knowledge Gemma 3 27B leads
Gemma 3 27B: 25.5 (#261), Mixtral 8x22B: 15.1 (#293)
| Benchmark | Gemma 3 27B | Mixtral 8x22B |
|---|---|---|
| GPQA Diamond | 47.7% | 34.1% |
| LMArena Expert | 1304 | 1113 |
| MMLU-Pro | — | 46% |
| Confabulations | 40.3% | — |
| Vectara Hallucination Rate | 7.4% | — |
| GPQA (HELM) | — | 33.4% |
| MMLU | — | 77.8% |
Multimodal Not comparable
Gemma 3 27B: 32.6 (#100), Mixtral 8x22B: —
| Benchmark | Gemma 3 27B | Mixtral 8x22B |
|---|---|---|
| LMArena Vision | 1164 | — |
| GeoBench | 52% | — |
Multilingual Gemma 3 27B leads
Gemma 3 27B: 46.9 (#155), Mixtral 8x22B: 32.8 (#255)
| Benchmark | Gemma 3 27B | Mixtral 8x22B |
|---|---|---|
| LMArena Non-English | 1334 | 1128 |
| LMArena Chinese | 1346 | 1116 |
| LMArena French | 1368 | 1166 |
| LMArena German | 1362 | 1141 |
| LMArena Japanese | 1287 | 1037 |
| LMArena Korean | 1308 | 1057 |
| LMArena Russian | 1349 | 1158 |
| LMArena Spanish | 1349 | 1151 |
Instruction Following Gemma 3 27B leads
Gemma 3 27B: 70.6 (#160), Mixtral 8x22B: 57.7 (#266)
| Benchmark | Gemma 3 27B | Mixtral 8x22B |
|---|---|---|
| LMArena Instruction Following | 1321 | 1147 |
| LiveBench Instruction Following | 74.9% | — |
| IFEval | — | 72.4% |
Long Context Mixtral 8x22B leads
Gemma 3 27B: 27.6 (#293), Mixtral 8x22B: 34.7 (#247)
| Benchmark | Gemma 3 27B | Mixtral 8x22B |
|---|---|---|
| LMArena Longer Query | 1333 | 1144 |
| Fiction.LiveBench | 33.3% | — |
Writing & Preference Gemma 3 27B leads
Gemma 3 27B: 52.5 (#168), Mixtral 8x22B: 36.9 (#262)
| Benchmark | Gemma 3 27B | Mixtral 8x22B |
|---|---|---|
| LMArena Text | 1358 | 1162 |
| LMArena Creative Writing | 1346 | 1141 |
| LMArena Multi-Turn | 1345 | 1130 |
| Short-Story Creative Writing | 79.9% | — |
| EQ-Bench Creative Writing | 1266 | — |
| WildBench | — | 71.1% |
| LiveBench Language | 34.6% | — |
Frequently asked questions
Is Gemma 3 27B better than Mixtral 8x22B?
Gemma 3 27B is the stronger model overall, scoring 30.8 to 27.1 on the Noometry Index.
Which is cheaper, Gemma 3 27B or Mixtral 8x22B?
Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; Mixtral 8x22B lists at $2 and $6.
Is Gemma 3 27B or Mixtral 8x22B better for coding?
Mixtral 8x22B scores higher on coding benchmarks: 24.2 versus 22.5 in the Noometry coding category.
Which has the bigger context window?
Gemma 3 27B does, with 131K tokens against 64K.
How many benchmarks do Gemma 3 27B and Mixtral 8x22B share?
21 benchmarks have published results for both models. Gemma 3 27B has 43 scored results on Noometry and Mixtral 8x22B has 34.